Mechanism of 25-Hydroxyvitamin D3 24-Hydroxylation: Incorporation of Oxygen- 18 into the 24 Position of 2 5 - H ydroxyvitamin D3t Thelma C. Madhok, Heinrich K. Schnoes, and Hector F. DeLuca*
ABSTRACT: The oxygen enzymatically inserted as a hydroxyl function by chick kidney homogenate into the 24 position of 25-hydroxyvitamin D3 to give 24,25-dihydroxyvitamin D3 is derived exclusively from lxOz. Therefore, like the 25-hy-
droxyvitamin D3- 1a-hydroxylase system, the 25-hydroxyvitamin D3-24-hydroxylase system is also a rnonooxygenase (“mixed-function oxidase”).
p r e v i o u s reports have established that vitamin D3 must be metabolically activated before it can manifest its physiologic functions (DeLuca, 1974; Kodicek, 1974). The activation of vitamin D3 primarily by the liver involves a hydroxylation reaction in the 25 position to produce 25-hydroxyvitamin D3 (25-OH-D3)’ (Horsting and DeLuca, 1969; Blunt et al., 1968; Ponchon et al., 1969). This hydroxylated derivative represents the major metabolite of vitamin D3 circulating in blood plasma, not only in experimental rats (Lund and DeLuca, 1966) and chickens (Ponchon et al., 1969; Lawson et a]., 1969), but also in man (Mawer et al., 1969; Gray et al., 197 1). The 25-OH-D3 is subsequently transported to the kidney where it is further hydroxylated on carbon I to yield 1,25-dihydroxyvitamin D3 ( I ,25-(0H)zD3) (Fraser and Kodicek, 1970; Holick et al., 197l;Grayetal., 1971; Normanetal., 1971)oroncarbon 24 to yield 24,25-dihydroxyvitamin D3 (24,25-(OH)zD3) (Holick et al., 1972) depending on physiological circumstances. The manner by which the formation of these hydroxylated metabolites are regulated is still unknown but it is known that the production of 24,25-(OH)2D3 is generally stimulated under conditions in which the production of 1,25-(OH)zD3 is suppressed (Boyle et a]., 1971; Garabedian et al., 1972; Tanaka and DeLuca, 1973). Furthermore, 1,25-(OH)2D3 stimulates the 24-hydroxylase (Tanaka and DeLuca, 1973; Tanaka et al., 1975). However, the function of 24,25-(OH)zD3 is still the subject of much debate, although it is generally agreed that 24,25-(OH)*D3 is a major metabolite of vitamin D in man and animals. Also the 24-hydroxylase also hydroxylates 1,25(OH)lD3 to form 1,24,25-trihydroxyvitaminD3 (1,24,25(OH)3D3) (Tanaka et al.. 1977; Kleiner-Bossaller and DeLuca, 1974). Regardless of the function of 24,25-(OH)zDj, the mechanism of 24-hydroxylation of vitamin D is of obvious importance. The 25-OH-D3- 1 -hydroxylase has been extensively studied and is in many respects similar to the monooxygenase of beef adrenal cortex mitochondria concerned with steroidogenesis
(Gray et al., 1972; Sih, 1969). Hydroxylation of 25-OH-D3 on carbon 1 takes place in isolated chick kidney mitochondria in the presence of the oxidizable substrate, molecular oxygen, and magnesium ions (Fraser and Kodicek, 1970; DeLuca, 1974; Tanaka and DeLuca, 1974; Gray et al., 1971,1972).The reaction also requires N A D P H (Ghazarian and DeLuca, 1974). Experiments with I8O2 have revealed clearly that the oxygen, which is placed in position 1 by the hydroxylase, arises exclusively from molecular oxygen (Ghazarian et al., 1973). These findings plus the recent demonstration of a requirement for cytochrome P450 (Ghazarian et al., 1974) and an ironsulfur protein (Ghazarian et al., 1974; Pedersen et al., 1976) in the mitochondria have clearly established the I a-hydroxylase as a monooxygenase (“mixed-function oxidase”). The 24-hydroxylase is also found in chick kidney mitochondria and also requires NADPH, magnesium ions, and molecular oxygen (Knutson and DeLuca, 1974). Unlike the I -hydroxylase, the 24-hydroxylase is not inhibited by carbon monoxide. Little is known further about this pivotal system in the metabolism of vitamin Dj. I n this paper it will be demonstrated that the oxygen enzymatically introduced as a hydroxyl function into the 24 position of 25-OH-D3 to yield 24R,25-(OH)zD3 is derived exclusively from molecular oxygen providing strong evidence that the 24-hydroxylase is B monooxygenase (“mixed-function oxidase”) rather than a dehydrogenase-hydratase system.
t From the Department of Biochemistry, College of Agricultural and Life Sciences, University of Wisconsin-Madison. Madison, Wisconsin 53706. Hereiced December 27. 1976. Supported by a grant from the National Institutes of Health No. AM-14881 and the Harry Steenbock Research Fund. Abbreviations used: 25-OH-D3. 25-hydroxyvitamin D3; 1,25(OH)2D3, I ,25-dihydroxyvitamin D3; 24,25-(OH)*D3, 24,25-dihydroxyvitarnin D3; 1,24,25-(OH)3D3, I ,24.25-trihydroxyvitamin D3; HPLC, high-pressure liquid chromatography.
’
2142
BIOCHEMISTRY. VOL.
16, NO. IO, 1977
Methods Animals. Seventeen 1 -day-old white Leghorn cockerel chicks were obtained from Northern Hatcheries (Beaver Dam. Wis.). They were maintained on a 3% calcium, vitamin Ddeficient purified soy protein diet for I O days (Omdahl et al., 1971) and were dosed with 1,25-(OH)zD? (250 ng per chick, intramuscularly) 24 and 48 h before sacrifice. Preparation of the Enzyme. The chicks were killed by cervical dislocation, and the kidneys were immediately removed and carefully separated from the adhering connective tissue. The kidneys were rinsed with ice-cold buffer (0.19 M sucrose-I 5 m M Tris-acetate (pH 7.4)-1.9 m M magnesium acetate), transferred to an ice-cold Potter-Elvehjem homogenizer fitted with a Teflon pestle, and homogenized in 2 volumes of the buffer. The homogenate was centrifuged a t 128g a t 4 “C for 5 min using a Lourdes centrifuge equipped with a 9RA rotor. The resulting supernatant was used for incubation. lncubation Conditions and Extraction of Samples. Four-
25-HYDROXYVITAMIN D3-24-HYDROXYLASE teen incubations, each 1.5 m L total volume, were performed in 5-mL serum bottles containing 19 mg of homogenate protein and 25 m M succinate. The incubation medium was described previously (Knutson and DeLuca, 1974). The bottles were sealed with serum bottle caps and connected through needle outlets to a vacuum train. The bottles were then evacuated repeatedly, followed by flushings with purified nitrogen passed over heated copper. Finally, 3.5 m L of 99.3 atom % l 8 0 2 (Bio-Rad Laboratories, Richmond, Calif.) was introduced into each of 14 incubation bottles using a gas-tight syringe. The hydroxylation reactions were initiated by introduction (with a Hamilton syringe) of 1.5 gg of 25-OH-[26,27-3H]D3 in 10 p L of 95% ethanol with a specific radioactivity of 247 703 cpm/gg. After 30 min of incubation at 37 OC and a t 120 oscillations/min, the reactions were terminated by the immediate transfer of the mixtures into a 500-mL separatory funnel containing 84 mL of 2:l CH30H-CHC13 mixture, and 63 mL of C H 3 0 H was then added. The reaction mixture was shaken and allowed to stand at room temperature for 1 h. A two-phase system a t 4 OC was produced by the addition of 35 m L of deionized water and 84 m L of CHC13. The lower phase was removed and the upper phase was reextracted twice with 84-mL portions of CHC13. The combined chloroform extracts contained 6.5 X lo6 cpm or 100% of the added radioactivity. Chromatography. The combined chloroform extracts were evaporated to dryness using a rotary evaporator a t 30 O C . The residue was dissolved in 500 g L of 65:35 CHCI~-Skellysolve B and applied to a 1 X 60 cm glass column containing 18 g of Sephadex LH-20 (Pharmacia Corp., Piscataway, N.J.), packed and eluted in the same solvent system (Holick and DeLuca, 1971). A total of 60 fractions (3.5 m L a t a flow rate of 0.6 mL/min) were collected and 5pL of each fraction was used for tritium counting (Figure la) in a Packard Model 3375 liquid scintillation spectrometer (Packard Instrument Co., Downers Grove, Ill.). The efficiency of counting was 35%. The relative position of the metabolites has been determined previously by chromatography of the pure metabolites under identical experimental conditions (Holick and DeLuca, 197 1; Holick et al., 1973). The 24R,25-(OH)zD3 peak region, fractions 30-40, was combined and evaporated to dryness using a rotary evaporator a t 30 O C to yield a total radioactivity of 465 666 cpm equivalent to 1.9 l g of 24,25-(OH)zD3 (based on 247 703 cpm/pg, the specific radioactivity of the substrate, 25-OH-D3). The residue was dissolved in 100 pL of 50:50 CHCI~-Skellysolve B and applied to a 1 X 60 cm glass column containing 18 g of Sephadex LH-20, packed and eluted in the same solvent system (Holick and DeLuca, 1971). A total of 80 fractions (3.5 m L at a flow rate of 0.6 mL/min) was collected and 20 pL of each fraction was used for tritium counting (Figure lb). The product peak region, fractions 48-74, was combined and evaporated to dryness using a rotary evaporator a t 30 O C to yield a total radioactivity of 433 782 cpm equivalent to 1.75 pg of 24,25-(oH)$3. The residue was dissolved in 100 p L of C H 3 0 H and applied to a 1 X 90 cm glass column packed with 20 g of Sephadex LH-20 in the same solvent. The column was eluted with CH3OH. A total of 60 fractions (2 mL a t a flow rate of 0.5 mL/min) was collected and 5 p L of each fraction was used for tritium counting (Figure IC). Peak fraction 32 (containing 125 280 cpm equivalent to 506 ng of product) was dried down under nitrogen at room temperature and applied to a 4 mm X 30 cm pPorasil column (Waters Associates) in a Du Pont 830LC apparatus fitted with a Waters U-6-K injection port. Elution was accomplished with 5% 2propanol in hexane a t 700 psi. A total of 12 fractions (2 m L
FRACTION NUMBER ( 3 . 5 ml )
FRACTION NUMBER (2.0rnl
FRACTION N U M B E R ( 3 . 5 ml)
RETENTION T I M E
rb'
(Minutes
Chromatographic profiles during the isolation and purification of 24-180-24,25-(OH)2-[26,27-3H]D3. (a) Sephadex LH-20 column ( I X 60 cm packed in and eluted with 6 5 : 3 5 CHC13-Skellysolve B). (b) Sephadex LH-20 column ( I X 60 cm packed in and eluted with 50:50 CHC13-Skellysolve B) profile of 24-180-24,25-(OH)2D3region isolated from the Sephadex LH-20 column (65:35) CHC13-Skellysolve B shown in a. (c) Sephadex LH-20 column ( 1 X 90 cm packed in and eluted with C H 3 0 H ) profile of 24-'80-24.25-(OH)2D3region isolated from the Sephadex LH-20 column (50:50)CHCI3-Skellysolve B shown in b. (d) FPorasil column (4 mm X 30 cm Waters; eluted with 5% 2-propanolhexane) profile of 24-I80-24,25-(OH)2D3 region isolated from the Sephadex LH-20 column (CH3OH) shown in c. FIGURE 1:
min-2 fraction-') was collected. The elution profile is shown in Figure Id. The peak fractions 8 and 9 were combined and dried down under nitrogen to yield a total residue of 202 ng of 24,25-(OH)zD3. This material was analyzed in the mass spectrometer (Associated Electrical Industries, Ltd., Manchester, England, Model MS-902) by direct probe inlet at 120- 130 "C and 70 eV ionization energy. Results Incubation of 25-OH-D3 with chick kidney homogenates has resulted in an overall enzymatic conversion of 9% to 24,25-(OH)zD3 (1.9 pg as calculated from the initial amount of substrate added which was 21 pg). The purification of the I8O metabolite depended heavily on liquid-gel partition chromatography (Holick and DeLuca, 1971) using Sephadex LH-20 suspended in double-distilled organic solvents as indicated throughout the procedures. As a final purification step, the more refined technique of high-pressure liquid chromatography was adopted. An excellent resolution (Figure l a ) is observed of 24,25-(OH)$3 from the unaltered, undegraded substrate 25-OH-D3, shown as the first radioactive peak. As observed from the same figure, a third radioactive metabolite was separated. This peak cochromatographed with crystalline 1,25-(OH)& on high-pressure liquid chromatography. The 1,25-(OH)2D3-biosynthesizing system, therefore, was still BIOCHEMISTRY. VOL.
16, NO. IO, 1977
2143
L u
f
1
> W
2%
t
a
4yI
271
2 W
311
nz
3m
A -’ - J c - ( 400
I
i t
400
F ICJURE 2 Mass spectrum of isoldted 24-’80-24,25-(OH)2-[26.27-3H]Dj The product (202 ng) from the pPorasil column (Figure I d ) \4a\ a n ~ l \ 7 c d by mass spectrometry Note especially the molecular ion at m/e 418 and the absence of m7/e 416 Note also important fragments a t m/e 27 I I36 ‘ind I18
present in these preparations. The recovery of metabolites from the various Sephadex columns was essentially quantitative. The product obtained after H P L C was sufficiently pure to allow mass spectral analysis. The isolation and identification of the 24,25-(OH)zD3 have been reported by Holick et al. ( 1 972). The mass spectrum of the natural dihydroxy metabolite was shown to exhibit a molecular ion peak a t m/e 416 and prominent fragment peaks at m/e 1 18, 136,257, and 27 I . The mass spectrum of the [‘*0]-24-hydroxy analogue of 25-OH-D3 (Figure 2) was compared with that of the natural metabolite (Holick et ai., 1972). The molecular ion peak a t m/e 418 and the fragment peaks at m/e 118, 136, 257, and 271 establish conclusively the incorporation of molecular oxygen as a hydroxyl function into the C-24 position of 25-OH-D3 to form 24,25-(oH)zD3. Note the absence of m/e 416 which would arise if H2O had provided the oxygen for the 24-hydroxyl function.
However, it was shown (Knutson and DeLuca, 1974) that, in contrast to the 1-hydroxylase, the 24-hydroxylase is not carbon monoxide sensitive which casts doubt on the involvement of cytochrome P450 in the 24-hydroxylation reaction. Yet mitochondria containing 24-hydroxylase have the same level of cytochrome P450 as the mitochondria containing 1 -hydroxylase (Ghazarian et al., 1974). It is possible that a cytochrome P450 is involved in the 24-hydroxylation but may be resistant to carbon monoxide inhibition. On the other hand, the 24-hydroxylase system may be a monooxygenase (“mixed-function oxidase”) not requiring cytochrome P450 as electron acceptor (Knutson and DeLuca, 1974). This question can only be resolved by reconstitution experiments similar to those carried out by Ghazarian et al. ( 1 974) and Pedersen et al. ( 1 976) for the I a-hydroxylase. However, the present investigation permits the exclusion of a mechanism involving dehydrogenation and hydration.
Discussion The results of this investigation have provided strong evidence that the chick kidney (24R)-hydroxylase belongs to the class of “mixed-function oxygenases.” This group of oxygenases catalyzes the consumption of one molecule of oxygen per molecule of substrate; one atom of the oxygen molecule appears in the product, the other is reduced to water by a hydrogen donor other than the substrate, in this case NADPH (Knutson and DeLuca, 1974). Microsomal oxygenases usually consist of at least a flavoprotein, a carbon monoxide-sensitive hemoprotein, cytochrome P450, a lipid component (Lu and Coon, 1968; Strobel et al., 1970) and may contain an iron-sulfur protein (Omura and Sato, 1964; Kimura and Suzuki, 1967). The 25-OH-DJ- I -hydroxylase is the first of the vitamin D enzymes that was shown to be a monooxygenase (“mixedfunction oxidase”). It was demonstrated with ‘ 8 0 2 (Ghazarian et al., 1973) that in this hydroxylation reaction all of the oxygen inserted by the chick kidney mitochondria into the l a position of 25-OH-D3 is derived from molecular oxygen. Furthermore, evidence was presented which demonstrated that NADPH and a carbon monoxide sensitive component are two specific intramitochondrial factors necessary for the conversion of 25OH-D3 to 1 ,25-(OH)2D3 (Ghazarian and DeLuca, 1974). Later experiments have established the hemoprotein component as cytochrome P450 and the participation of an iron-sulfur protein called renal ferredoxin (Ghazarian et al., 1974; Pedersen et al., 1976). That the 24-hydroxylase also incorporates molecular I8O into 25-OH-D3 to form 24R,25-(OH)~D3suggests that this system is also a monooxygenase (“mixed-function oxidase”).
We express our sincerest appreciation to Dr. Robert H . Burris and Mr. David Emerich for their technical help in the handling of commercial l8O2.
2144
BIOCHEMISTRY. VOL.
16, N O . 10, 1 9 7 7
Acknowledgment
References Blunt, J. W., DeLuca, H. F., and Schnoes, H. K. (1968), Biochemistry 7 , 33 17. Boyle, I . T., Gray, R. W., and DeLuca, H. F. (1971), Proc. Natl. Acad. Sci. U.S.A. 68, 2 13 I . DeLuca, H. F. (1969), Fed. Proc.. Fed. A m . Soc. E x p . Biol. 28, 1678. DeLuca, H . F. ( 1 974), Fed. Proc., Fed. A m . Soc. E x p . Biol. 33, 221. Fraser. D. R., and Kodicek, E. ( 1 970), Nature (London)222, 171. Garabedian, M., Holick, M . F., and DeLuca, H. F. (1972). Proc. Natl. Acad. Sci. U.S.A. 69, 1673. Ghazarian, J. G., and DeLuca, H. F. (1974), Arch. Biochenz. Biophys. 160, 63. Ghazarian, J . G., Jefcoate, C. R., Knutson, J . C., OrmeJohnson, W. H., and DeLuca, H. F. (l974), J . Biol. Chem. 249, 3026. G h a ~ a r i a nJ. , G., Schnoes, H. K., and DeLuca, H. F. ( 1 973), Biochemistry 12, 2555. Gray, R., Boyle, I. T., and DeLuca, H. F. (1971). Science 172, 1232. Gray, R. W., Omdahl, J. L., Ghazarian, J . G., and DeLuca, H. F. (1972), J . Biol. Chem. 247, 7528. Holick, M., and DeLuca, H. F. (1971). J . Lipid Ke.5. 12. 460. Holick, M. F., Schnoes, H. K., DeLuca, H . F., Gray, R. W..
EXCHANGE OF PHOSPHATIDYLCHOLINE
Boyle, I. T., and Suda, T. (1 972), Biochemistry 1 1 , 425 I . Holick, M. F., Schnoes, H. K., DeLuca, H. F., Suda, T., and Cousins, R. J. (1971), Biochemistry I O , 2799. Holick, M. F., Semmler, E. J., Schnoes, H. K., and DeLuca, H. F. (1973), Science 180, 190. Horsting, M., and DeLuca, H . F. (1969), Biochem. Biophys. Res. Commun. 36, 251. Kimura, T., and Suzuki, K. (1967), J . Biol. Chem. 242, 485. Kleiner-Bossaller, A., and DeLuca, H . F. (1974), Biochim. Biophys. Acta 338, 489. Knutson, J. C., and DeLuca, H. F. (1974). Biochemistry 13, 1543. Kodicek, E. (1974), Lancet 1 , 325. Lawson, D. E. M., Wilson, P. W., and Kodicek, E. (1969), Nature (London) 222, 17 1. Lu, A. Y . H., and Coon, M. J . (1968), J . Biol. Chem. 243, 1331. Lund, J., and DeLuca, H . F. (1966), J . Lipid Res. 7, 739.
Mawer, E. B., Lumb, G. A., and Stanbury, S. W. (1969), Nature (London) 222, 482. Omdahl, J. L., Holick, M. F., Suda, T., Tanaka, Y., and DeLuca, H. F. (1971), Biochemistry IO, 2935. Omura, T., and Sato, R. (1964), J . Biol. Chem. 239, 2379. Pedersen, J. I., Ghazarian, J. G., Orme-Johnson, N. R., and DeLuca, H. F. (1976), J. Biol. Chem. 251, 3933. Ponchon, G., Kennan, A. L., and DeLuca, H. F. (1969), J . Clin. Invest. 48, 2032. Sih, C. (1969), Science 163, 1297. Strobel, H. W., Lu, A. Y . H., Heidema, J., and Coon, M. J. (1970), J . Biol. Chem. 245, 4851. Tanaka, Y., Castillo, L. C., DeLuca, H. F., and Ikekawa, I . (l977), J . Biol. Chem. 252, 1421. Tanaka, Y., and DeLuca, H. F. (1973), Science 183, 1198. Tanaka, Y . ,and DeLuca, H. F. (l974), Proc. Natl. Acad. Sci. U.S.A. 7 1 , 1040. Tanaka, Y . ,Lorenc, R. S., and DeLuca, H. F. (1 975), Arch. Biochem. Biophys. 171, 521.
Protein-Catalyzed Exchange of Phosphatidylcholine between Sonicated Liposomes and Multilamellar Vesicles? Paul E. DiCorletot and Donald B. Zilversmit*,§
ABSTRACT:Phospholipid exchange protein from beef heart or beef liver does not catalyze the transfer of phosphatidylcholine from multilamellar vesicles of phosphatidylcholine. Certain combinations of phospholipids, however, do yield multilamellar vesicles that will exchange phosphatidylcholine with liposomes in the presence of exchange protein. Multilamellar vesicles of phosphatidylcho1ine:phosphatidylethano1amine:cardiolipin (70:25:5, mol %) can be used in place of mitochondria or erythrocyte ghosts as an improved acceptor particle in the study of liposome structure with phospholipid exchange proteins. These multilamellar vesicles act as a well-defined reservoir of unlabeled phosphatidylcholine with 7% exchangeable phospholipid. When the distribution of phosphatidylcholine in liposomes is studied by the exchange
protein technique, results can be influenced by the choice of phospholipid acceptor particle. With mitochondria as acceptor particle, the percentage of phosphatidylcholine in the outer monolayer of a liposome appears to be 60%, whereas a value of 70% is obtained when multilamellar vesicles are the acceptor. The discrepancy can be explained by a heterogeneity in liposomes prepared by sonication. A size-dependent fusion or adsorption process occurs between liposomes and mitochondria; the very small liposomal vesicles, obtained by gel filtration, combine nearly quantitatively with the natural membrane. This phenomenon is not seen with multilamellar vesicles. Thus by using multilamellar vesicles one obtains a less biased estimate of phospholipid distribution between inner and outer layers of liposomes.
R o t e i n s that catalyze the transfer of phospholipid molecules between membranes have been isolated and characterized from several sources (for review, see Zilversmit and Hughes, 1976). Recently, these phospholipid exchange proteins have been employed in several laboratories as membrane probes in the study of phospholipid distribution and motion in liposomes-
sonicated, unilamellar vesicles (Johnson et al., 1975; Rothman and Dawidowicz, 1975) and natural membranes (Bloj and Zilversmit, 1976; Rothman et a]., 1976; Rousselet et al., 1976). The value of the technique lies in the high degree of sensitivity of radioisotope monitoring and the apparently minimal perturbation of the bilayer. The transposition (flip-flop) rate of PC' in liposomes has been determined with this technique in two laboratories by the use of radioactive PC liposomes as donor particles and mitochondria (Johnson et al., 1975) or erythrocyte ghosts (Rothman and Dawidowicz, 1975) as acceptor particles. The kinetics of catalyzed replacement of donor
+ From the Division of Nutritional Sciences and Section of Biochemistry, Molecular and Cell Biology, Division of Biological Sciences, Cornell University, Ithaca, New York 14853. ReceiuedSeptember 16, 1976. This research was supported in part by Public Health Service Research Grant HL 10940 from the National Heart and Lung Institute, United States Public Health Service. 1 National Institutes of Health predoctoral trainee supported by Training Grant No. G M 00824 from the National Institute of General Medical Sciences, United States Public Health Service. 5 Career Investigator of the American Heart Association.
I Abbreviations used: PC, phosphatidylcholine; C L , cardiolipin; PI, phosphatidylinositol; PE, phosphatidylethanolamine; PA, phosphatidic acid; Pi, inorganic phosphate; Tris, tris(hydroxymethy1)aminomethane; EDTA, ethylenediaminetetraacetic acid.
BIOCHEMISTRY, VOL.
16,
NO.
I O , 1977
2145
